On 2 February 2009, the German authorities have notified through the Rapid Alert System for Food and Feed the migration of 4-methylbenzophenone from packaging into certain cereal products, in a concentration of 798 micrograms/kg. The Belgian Authorities have also provided data later in February, reporting concentrations of 4-methylbenzophenone in cereals up to 3729 mu g/kg. In accordance with Article 29 (1) (a) of Regulation (EC) No 178/2002, the European Commission has asked the European Food Safety Authority to evaluate if the substance 4-methylbenzophenone would be covered by the TDI on benzophenone and hydroxybenzophenone and evaluate the risk of the presence of 4-methylbenzophenone found in cereals by 3 March 2009. In addition the Commission has asked EFSA to re-asses the TDI on benzophenone and hydroxybenzophenone in view of the new toxicological studies available, by end of May 2009. Concerning the first request on 4-methylbenzophenone, EFSA has informed the Commission that due to such a short timeline, it is only able to publish an EFSA statement on the issue rather than a CEF Panel opinion. Concerning the second request on the TDI on benzophenone and hydroxybenzophenone, the CEF Panel is expected to adopt an opinion by the end of May 2009. 4-Methylbenzophenone is used as a photo-initiator of UV-cured printing inks and lacquers applied on the surface of the packages, mainly cardboard boxes. 4-Methylbenzophenone may be used in combination with or as replacement of benzophenone. Given their volatility, both substances may migrate into the package and contaminate even solid foodstuffs. In the literature, there is very little information useful for the toxicological assessment of 4-methylbenzophenone itself. Much more information is available on benzophenone, which has a very similar structure. There is insufficient direct evidence available on the genotoxicity of 4-methylbenzophenone. However, based on structural considerations and experimental results on the structurally related benzophenone, it can be concluded that 4-methylbenzophenone does not raise concern for genotoxicity. 4-Methylbenzophenone is expected to be metabolised by the same metabolic pathways as benzophenone, with the addition of oxidation of the 4-methyl group to the corresponding alcohol and further oxidation to the carboxylic acid with its glycine and glucuronide conjugates. Like benzophenone, 4-methylbenzophenone is expected to be a non genotoxic carcinogen. Due to the lack of toxicity data for 4-methylbenzophenone and to the specific metabolic pathway, it is currently not scientifically sound to include 4-methylbenzophenone into the group-TDI for benzophenone and hydroxybenzophenone. Therefore at present a TDI cannot be derived for 4-methylbenzophenone. For the purpose of addressing this urgent request for advice to risk managers, the Margin of Exposure (MoE) approach is proposed. An uncertainty factor of two is considered appropriate for read-across from benzophenone. Benzophenone has been studied in subchronic (90-day) and chronic studies. In the latter, liver tumours were observed at a LOAEL of 15 mg/kg b.w. per day. In a two-generation study with benzophenone, livers of parental animals showed adverse effects, which could be related to the carcinogenicity in the chronic assay. From this study a LOAEL of 6 mg/kg b.w. per day could be derived. This LOAEL is used as a basis for calculation of the MoE for 4-methylbenzophenone. A factor of 100 for inter-and intraspecies differences in sensitivity, a factor of 3 for use of a LOAEL (instead of a NOAEL) and a factor of 2 for read-across from benzophenone to 4-methylbenzophenone are applied. Hence the estimated MoE should be greater than 600. The MoE is calculated by dividing the LOAEL (6 mg/kg b.w. per day) by the estimated exposure in each case. Based on analytical results from 17 samples of breakfast cereals as provided by the Belgian Authorities (AFSCA, 2009), exposure estimates to 4-methylbenzophenone were derived for children and adults according to a conservative and to a highly conservative scenario. Both scenarios are based on the highest consumption of breakfast cereals as registered in Ireland: 2.45 g/kg b.w. per day for children and 0.96 g/kg b.w. per day for adults (Table 2). The conservative scenario assumes average concentration of 4-methylbenzophenone as calculated from the results on the 17 samples from the Belgian market whilst the highly conservative scenario assumes the maximum concentration found in these samples. For children the conservative scenario leads to a dietary exposure via breakfast cereals of 2 mu g 4-methylbenzophenone/kg b.w. per day. The highly conservative scenario for children leads to a dietary exposure via breakfast cereals of 13.2 mu g 4-methylbenzophenone/kg b.w. per day. For adults the conservative and the highly conservative scenarios lead to dietary exposures via breakfast cereals of 0.79 and 5.2 mu g of 4-methylbenzophenone/kg b.w. per day, respectively. In the absence of adequate data, a rough estimate of hypothetical dietary exposure to 4-methylbenzophenone from other sources was calculated, based on the consumption of food that may contain benzophenone as a proxy, given that these two substances are used for a similar purpose. For that purpose a survey on benzophenone (FSA, 2006) was used assuming the 4-methylbenzophenone was used instead. This resulted in hypothetical dietary exposures of 1 mu g 4-methylbenzophenone/kg b.w. per day for adults and 2 mu g 4-methylbenzophenone/kg b.w. per day for children. The hypothetical overall exposure from all potentially contaminated foods has been estimated by combining exposure from all potentially contaminated foods (breakfast cereals + other foods). The conservative scenario leads to an overall exposure of 1.79 mu g 4-methylbenzophenone/kg b.w. per day for adults and 4 mu g 4-methylbenzophenone/kg b.w. per day for children The highly conservative scenario leads to a hypothetical overall estimated exposure from all potentially contaminated foods of 6.2 mu g 4-methylbenzophenone/kg b.w. per day for adults and 15.2 mu g 4-methylbenzophenone/kg b.w. per day for children. Based on the LOAEL of benzophenone of 6 mg/kg b.w. per day and the hypothetical overall exposure from all foods, Margins of Exposure could be estimated as follows: For adults: 3351 and 968 for conservative and highly conservative scenarios respectively. For children: 1500 and 395 for conservative and highly conservative scenarios respectively. Thus for adults, the estimated exposure is unlikely to lead to a health concern, since the estimated MoE is higher than 600. For children, the estimated exposure based on a conservative scenario (high consumption of breakfast cereals, average concentration of 4-methylbenzophenone) is also unlikely to pose a health concern. However, for children, based on the highly conservative scenario (high consumption of breakfast cereals, highest concentration of 4-methylbenzophenone), the estimated MoE is below 600. Therefore a health concern cannot be excluded in this case. Based on the limited exposure data available and applying knowledge on the toxicity of a similar substance, benzophenone, EFSA concludes that short term consumption of contaminated breakfast cereals should not pose a risk to most people. However, if the use of 4-methylbenzophenone is to be continued, more data on occurrence of the substance in foods should be provided as well as appropriate toxicity data corresponding to the level of exposure for a full risk assessment.
